A review of robotic grasp detection technology

Abstract

In order to complete many complex operations and attain more general-purpose utility, robotic grasp is a necessary skill to master. As the most common essential action of robots in factory and daily life environments, robotic autonomous grasping has a wide range of application prospects and has received much attention from researchers in the past decade. However, the accurate grasp of arbitrary objects in unstructured environments is still a research challenge that has not yet been completely overcome. A complete robotic grasp system usually involves three aspects: grasp detection, grasp planning, and control subsystem. As the first step, identifying the location of the object and generating the grasp pose is the premise of successful grasp, which is conducive to planning the subsequent grasp path and the realization of the entire grasp action. Therefore, this paper conducts a literature review focusing on grasp detection technology and concludes two significant aspects: the analytic and data-driven methods. According to the previous grasp experience of the target object, this paper divides the data-driven methods into the grasp of known and unknown objects. Then it describes in detail the typical grasp detection methods and related characteristics of each classification in the grasp of unknown objects. Finally, current research status and potential research directions in this field are discussed to provide some reference for related research.

Automatic loading method for robot guided by 3D vision for scattered and stacked shoe soles

“…Therefore, we investigated cases of using industrial robots to complete grabbing and loading tasks in industrial application scenarios. After analyzing 17 studies on robot grabbing success rates surveyed in the review literature [30], we found that the average robot grabbing success rate reported in these studies was 82.21%. In contrast, our proposed method achieved a success rate of 97.08%, surpassing the average level.…”

Effective grasp detection method based on Swin transformer

“…This technology is extensively employed in a variety of scenarios, including warehousing logistics, industrial automation, etc. 1,2 In unstructured environments, objects exhibit uncertainty and location randomness, which pose challenges to the effective execution of grasping tasks. The characteristics, including type, size, shape, and other attributes, of objects contribute to this uncertainty, whereas the random positioning of objects implies that their spatial coordinates within the robot's operational scope remain unfixed, potentially appearing anywhere within its reach.…”

“…Our work in this paper is as follows: (1) we designed the Swin-transNet grasp detection framework, highlighting our systematic integration of established technologies to mitigate the challenges associated with object uncertainty, random positions, and perspective differences in grasp detection tasks. (2) We studied the operating mechanism of each component within the algorithm framework, and analyzed the role of Swin transformer and the decoupled head and attention mechanism in the grasp detection process. (3) We conducted comprehensive experiments to validate the effectiveness of Swin-transNet.…”

Design of a Wheelchair-Mounted Robotic Arm for Feeding Assistance of Upper-Limb Impaired Patients

“…Additionally, future work will delve into control aspects, exploring impedance and admittance control for optimal robot-environment interaction, as proposed in works such as [29]. Attention will also be directed towards active learning of collision distance functionality, as suggested in [30] and [31].Furthermore, we aim to explore different end-effectors, such as grippers capable of manipulating not only food but also a wide range of objects, as proposed by Fu et al [32], Dong and Zhang [33], Carbone et al [34], and Yao et al [35]. Their collective work offers valuable insights into control strategies, motion planning, flexible grasping, grasp detection technology, and optimal design of robotic mechanisms, thereby opening multiple avenues for further enhancement of this research.…”